Remaking Work, Remaking Space: Spaces of Production and Accumulation in the Reconstruction of American Capitalism, 1865-1920

© 2002 Editorial Board of Antipode. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden,MA 02148, USA

Remaking Work, Remaking Space:Spaces of Production and

Accumulation in the Reconstructionof American Capitalism, 1865–1920

Jason W MooreDepartment of Geography, University of California, Berkeley, CA, US;

e-mail: [email protected]

The era of US capitalist development between 1865 and 1920 offers a good oppor-tunity to analyze the relational nature of social change at multiple scales preciselybecause it was a time of transition, for US and world capitalism alike. Existing accountsof the transition to monopoly capitalism in the US have focused on one or twogeographical scales, such as the national economy or the shop floor. In this literature,scales are essentially treated as “containers” within which social change occurs. Thepossibility that the containers themselves may be fundamentally altered is not addressed.In contrast, this paper views labor process transformations, and transformations of thesocial division of labor, as dialectically bound. In particular, I seek to explain how theAmerican transition to monopoly capitalism shaped, and was shaped by, class conflictand competitive pressures at multiple scales—the shop floor, the region, and thenational and global divisions of labor.

IntroductionIt is now widely recognized by geographers—and, increasingly othersocial scientists—that capitalism functions at and produces multiplegeographical scales (Brenner 1999; Harvey 2000; Swyngedouw 1997).Unfortunately, recognizing the existence of multiple scales hardlyclarifies matters. There remains the question of the relevant scale—orthe decisive relationship between scales—for the problem at hand.(This question is all the more vexing for the fact that both spatialconfigurations and the scales themselves are ever-shifting.) This is aparticularly big problem for the literature on capitalist development,where scholars have traditionally assumed a decisive scale of action,and generalized from that basis. For instance, the debate over capital-ist origins pits a “productionist” camp, emphasizing the productionunit, against a “circulationist” camp prioritizing world-scale forces(Brenner 1977; Wallerstein 1974). This paper seeks to undermine suchone-sided determinations by analyzing the spatial moment ofAmerican capitalism’s transformation between the end of the CivilWar and the close of World War I.

Remaking Work, Remaking Space 177

Existing accounts of the transition to monopoly capitalism in the UShave focused on one or two scales—the national economy or the rela-tionship between shop-floor struggles and the national political econ-omy (see, respectively, Baran and Sweezy 1966; Gordon, Edwards andReich 1982). In this literature, scales are essentially treated as “con-tainers.” Social change occurs within scales, and the relations betweenthem may shift. The possibility that “the containers themselves [maybe] fundamentally recast” (Smith 2000) is not addressed. I propose analternative that challenges such containerization of scale, and puts thepolitical economy of the production of space—emphasizing the mut-ability of scale—at the center. In particular, I seek to explain how theAmerican transition to monopoly capitalism shaped and was shapedby class conflict and competitive pressures at multiple scales—theshop floor, the region, and the national and global divisions of labor.

The era of US capitalist development between 1865 and 1920 offersa particularly good opportunity to analyze the relational nature ofsocial change at multiple geographical scales precisely because it wasa time of transition, for the US and world-system alike. Put abstractly,social systems during periods of stability may compensate for disrup-tions occurring at any given scale. During periods of transition, how-ever, changes at any one scale exert a much more powerful—andpotentially disruptive—effect on social relations at other scales. Inthese times, the mutually relational nature of social relations at mul-tiple scales becomes particularly evident. Such periods of crisis andchange illuminate the fault lines of the old sociospatial order, andtheir realignment into a new one.

The Civil War resolved the explosive contradictions posed by thecoexistence of plantation and industrial capitalism in one country. TheAmerican South was liberated from British hegemony and the ques-tion of Western settlement was decided in favor of independent small-holding agriculture. The victory of Northern industrial and Westernagrarian interests over the Southern planter regime allowed the US topursue “core” rather than “peripheral” capitalist development—thatis, a course of development that resembled northwestern Europe ratherthan Latin America.1 This allowed, for the first time, the geographicalintegration of a far-flung continental economy with an enormouslyrich resource base (Agnew 1987; Arrighi 1999; Post 1995).

At the same time, new contradictions moved to center stage. The con-solidation of a political regime favorable to core capitalist develop-ment opened the door to rapid economic growth and the consolidationof antebellum territorial conquests. In the half-century after 1865,American capitalism effected a transition from a quasi-national-scalesystem dominated by regional interenterprise competition, interregionalpolitical divisions, and small- to medium-scale industry to a nationallyunified monopoly capitalist order dominated by “giant corporations”

competing on a national, continental, and even world scale. Increas-ingly, industrial production was mass production (Baran and Sweezy1966; Hounshell 1984). The US working class grew a whopping 700%(Kinder and Hilgemann 1978:117). Manufacturing value-addedincreased seventeen-fold (Blackford 1988:48). Seven times as muchrailroad track was laid between 1860 and 1873 as in the previous threedecades (Agnew 1987:49). Such growth and expansion contained twomajor contradictions. First, individual capitals were quickly exposedto the promises and perils of intensified competition in a national mar-ket. Second, capitalists soon found themselves facing an increasinglylarge and militant industrial working class.

These contradictions drove mutually relational transformations onseveral scales. I advance three propositions on this subject.

First, the shop floor is a geographical scale whose spatiality reflectsand embodies class relations. The later 19th century witnessed a tran-sition from the workshop and simple factory organization to the modernindustrial plant. The rise of the giant corporation and the rise of thegiant factory are dialectically bound. Increasingly subject to consciousplanning and aided by technological innovations in concrete, steel,and electricity, owners now designed plants to fragment workers spa-tially and to replace workers’ cooperation with managerial coordination.Without such reconstruction of the workplace “built environment,”the success of the scientific management movement is scarcely con-ceivable. Prior to scientific management, skilled workers wieldedconsiderable autonomy and power on the shop floor (Montgomery1979, 1987). With the advent of continuous flow production methodsand mass production at the turn of the century, the layout of the shopfloor changed. Firms restructured the shop floor so that managers,rather than workers, increasingly set the pace of work. Not only werework teams split up and reorganized, but the skilled workers whoremained were often geographically separated from the semi- andunskilled workers (Pietrykowski 1995). This was crucial, from man-agement’s perspective, because skilled workers had historically beenthe most militant.

Second, industrial capital for the first time in world history becamefairly mobile, thanks largely to new transportation and communicationstechnologies such as the railroad, the steamship, and the telegraph.Indeed, the “time–space compression” that attracted such notice inthe late 20th century has its origins in these 19th-century innovations(Arrighi 1999; Harvey 1989). Some capitalists had always been mobile;one can think of the Dutch VOC (East India Company) in the 17thcentury or city-state finance capitalists even earlier (Arrighi 1994).What was profoundly new in the later 19th century was the new-found mobility of industrial capitalists. This new mobility permittedthe relocation of production facilities from the largest cities to industrial

178 Antipode

suburbs as a strategic response to workers’ power (Ashton 1978;Gordon 1978).

Third, and in many ways most fundamentally, the rise of the giantfactory and the giant corporation was predicated on the creation of thegiant economy. Never before had such an expansive “national” econ-omy assumed the leading role in the capitalist world-system. It is moreuseful to think of American capitalism as “continental” rather thannational. Always before, geographical “bigness” had been a positiveliability—France and Spain, for instance, fought losing battles throughthe early modern era against their more compact British and Dutchadversaries. It was American capitalism’s great innovation to weldtogether the new social form of the giant firm with the new spatialforms of the giant factory and the giant continental economy.

The creation of such an economy depended on a deeper and widerdivision of labor between town and country, within (so-called) nationaleconomies and on a world scale.2 After midcentury, urbanization pro-ceeded at a fantastic pace throughout the advanced capitalist world,brought about by two main developments: (1) the globalization of agri-cultural markets and the shift towards more specialized and capitalintensive forms of agriculture, characterized by successive “agriculturalrevolutions,” leading to successive waves of agrarian dispossessionand the destruction of rural handicrafts; and (2) the diffusion of large-scale industry under the impetus of larger national markets and thetechnologies of the “second industrial revolution” (Barraclough 1967:45–55; Hobsbawm 1987:342–343; Thompson 1968). “In short, theworkers were being gathered into factories and the factories concen-trated in industrial towns and urban areas” (Barraclough 1967:52). Ofcourse, many dislocated peasants could not be accommodated withintheir respective national economies, especially in Southern and EasternEurope. The globalization of agriculture set in train during the boomyears of the 1840s through the 1870s led to a renewed phase ofworldwide primitive accumulation in subsequent decades, driving tensof millions of peasants from the land, many of whom found their wayto US cities (Moore 2000; Wolf 1982). It was precisely these workerswho provided the crucial ingredient for America’s remarkable indus-trial expansion and its ascent to world power.

Thus, the dialectical antagonism of town and country assumed cen-tral importance throughout this period, for the local organization ofproduction no less than global structures of accumulation. At mid-century, Britain supported the creation of a global agricultural marketthrough its repeal of the Corn Laws in 1846 and worldwide investmentin transportation infrastructure. Vast new areas of agricultural productionwere integrated into the world market. The crisis of overproductionthat inevitably resulted when the British-led boom went bust in 1873led to a much more thoroughgoing capitalist transformation of


agriculture in these areas. On the one hand, such capitalist transform-ation meant dispossessed peasants, enlarging the national—and, to acertain extent, the global—reserve army of labor. On the other hand,it allowed for sustained industrial growth and capital accumulation bycreating rural markets for producer and consumer goods, which farmhouseholds could no longer do without or produce on their own. Evenmore basically, the subordination of agriculture to the “universalcompetition” of industrial capitalism created new imperatives forincreased productivity, thereby effecting a global “revolution in themethods of feeding an industrialized and urbanized population”(Barraclough 1967:50).

These global transformations were the decisive precondition for thereorganization of the industrial labor process that in time allowed forsharp increases in productivity. To this issue we will now turn.

Labor Process Transformation and the New Shop Floor GeographyBy the later 19th century, the giant factory emerged as the spatialexpression of American monopoly capitalism. The rise of the giantfactory was the necessary corollary to the rise of the era’s ascendantgiant corporations. Far from an automatic process deriving from tech-nological imperative, the transition from the artisanal workshop andsimple factory to the modern industrial plant was driven by the grow-ing concentration of workers—and deepening class conflict—at thepoint of production.

The American working class grew 164% between 1870 and 1913,more than double the rate of its chief rival for world economic leader-ship, Germany (calculated from Maddison 1995:246). Between 1860and 1900, the number of US manufacturing workers grew fourfold,from 1.5 to 6 million (ASHP 1992:12). In the three decades after 1860,“average shop size jumped from six [workers] to over twenty,” so thatby 1899 40% of manufacturing was taking place in factories (Laurie1989:116; Robinson and Briggs 1991:623). The largest factories be-came truly enormous, increasing “from roughly 1500 workers perestablishment [in the later 19th century] to a range of 20,000 to 60,000during the 1920s” (Gordon, Edwards and Reich 1982:116, 133). Suchconcentration encouraged unionization, which posed even greaterproblems for employers. After the setbacks of the 1890s, the labormovement grew from some 548,000 workers to 2.5 million between1900 and 1909 (Webb 1911:606).

The growing scale of production in the decades following the CivilWar was not at first accompanied by a transformation of industrialorganization. The primary factor behind manufacturing growth re-mained increasing inputs of labor relative to fixed capital. Factory pro-duction in leading industries such as metalworking was still controlled

180 Antipode

by skilled workers operating either as “inside contractors” or as fore-men (Clawson 1980:126–144; Montgomery 1979). Indeed, the numberof skilled operatives doubled from one to two million and their share ofthe workforce held steady between 1870 and 1890 (Shefter 1986:202).

The sharp increase in the spatial concentration of workers provokeda “control crisis” beginning in the 1880s (Edwards 1979:51). Themethods of labor supervision practiced in the small- and medium-sized shops of competitive industrial capitalism (ca. 1800–1870) wereincreasingly unable to cope with the growing scale of production. “Thiscrisis emerged from the contradiction between the firm’s increasingneed for control [under conditions of intensified interfirm competitionin a national market] … and its diminishing ability to maintain control… The need for control was a concomitant of the firm’s continuinggrowth” (Edwards 1979:51).

As a result, industrial capitalists were compelled to reorganize theproduction process. There is, of course, a vast literature on labor pro-cess rationalization that focuses more or less narrowly on the socialrelations of production (see Braverman 1974; Clawson 1980; Edwards1979; Gordon, Edwards and Reich 1982; Marglin 1974). In RichardC. Edwards’ (1979) formulation, the control crisis of the 1880s ledcapitalists to substitute structural control for hierarchical control, whichdepended on the active and personalized intervention of foremen andmanagers. The form of structural control that prevailed in the later19th century was technical control, “the classic image” of which is theassembly line, whereby managerial control is embedded in thematerial structure of work (Edwards 1979:20–21, chapter 8). Edwardsgets it partly right. Capital asserts its structural-technical control notonly through the organization of work—itself a spatial process—butalso through factory design itself, an issue to which we shall turnmomentarily.

Technical control rested upon the foundation of the scientific man-agement movement. To assert this kind of control, managers needed toundermine skilled workers’ power and to reorganize the labor processby separating conception from execution in the overall production pro-cess. This was Frederick Winslow Taylor’s contribution. Taylorism’s3

three central principles run as follows (Braverman 1974:112–121).First, managers should study every aspect of the production process sothat they, rather than workers, are in control. In this way, “the laborprocess is to be rendered independent of craft, tradition, and theworkers’ knowledge” (Braverman 1974:113). The second principle isthe “separation of conception from execution”: a strict division of laborshould be imposed on the production process so that “all possiblebrain work [is] removed from the shop floor” (Taylor, quoted inBraverman 1974:114; emphasis in Taylor). Managers increasingly dothe “brain work,” directing the workers with precise instructions. The


third principle is the assertion of managerial power on the shop floor.The application of the first two principles involved management’s studyof the labor process and the imposition of a strict division betweenmental and manual labor. The application of the third principle entailedthe specific and ongoing direction and continuing reorganization ofevery step in the production process so that progressively greater powerwas transferred to management in the interests of producing pro-gressively greater amounts of surplus value.

The application of these principles was uneven across industrial sec-tors and even between firms within the same sector (Edwards 1979;Storper and Walker 1989:154–182). Rather than focus on this uneven-ness, however, I wish to draw attention to how Taylorism dovetailedwith changes in the geography of the factory and shop floor. Thisallows us to treat the shop floor as a geographical scale that is madeand remade through the process of class conflict as well as intercap-italist competition (Sayer and Walker 1992:119–121; Storper and Walker1989:165–166). It is therefore useful to conceptualize the scientificmanagement movement as a major element of capital’s strategy totake advantage of shifts in the broader social division of labor to re-construct the scale of the shop floor.

The era of scientific management shaped and was shaped by theemergence of a new factory form. While factories certainly existedbefore the 1870s (Haraven and Langenbach 1978; Hounshell 1984),only after the Civil War did the modern industrial plant emerge as an“identifiable architectural form” in the US (Nelson [1975] 1995:11).The new factory was made possible by technological innovations instructural steel and reinforced concrete, but especially by advances inelectrical power. By the later 1880s, engineers had pioneered theindustrial application of electrical power, permitting two developments.First, electricity relieved industry of its strong dependence on riversfor power.4 As we shall see in the next section, this would have pro-found implications for capital mobility in the ensuing decades. Second,the existence of a “central generating source” made possible new fac-tory designs, “including decentralization, dispersion of work areas, andassembly lines” (Nelson 1995:11; Trachtenberg 1982:55–57). Factorieswere increasingly subject to conscious planning; they were no longermerely a “place to store machinery” (Nelson 1995:12, 15, 22). Gonewas the “hollow square” factory. In its place, engineers devised a num-ber of alternatives. “In general the trend was from the hollow squareto the other forms, from the multistory shop to the single-story plant,and from the ‘group’ to the ‘output’ (emphasizing product ratherthan machine function) plan” (Nelson 1995:21). Factory planners nowconsidered not just the general type of production, but “the flow ofwork between departments” (Nelson 1995:23; emphasis added). Tech-nical control was increasingly expressed in factory design. Rather than

182 Antipode

allowing workers to consciously organize the cooperation necessary forproduction, capital remade factories to ensure that managers couldnow assume the coordinating function and use this power to fragmentworkers at the point(s) of production. This became especially evidentby the early 20th century, when capital began building larger plantsaway from the epicenters of working class conflict in the big cities:

Large corporations … changed plant design significantly [in the early20th century]. As plants grew larger, firms gradually abandoned theclassic nineteenth-century model of the single open shed. A numberof important technical innovations [architectural expressions oftechnical control], such as reinforced concrete for construction andelectric power for traveling cranes, railroads, and other handlingequipment, permitted more flexible plant design. After 1895–1900,… most modern factories “consisted of a series of interrelated build-ings rather than a single large structure.” For example, foundrieswere located in separate structures, isolated from the main assemblyareas. Even within the general flow of assembly production, plantactivities were fragmented among disparate shops and structures.(Gordon, Reich and Edwards 1982:139; internal quotation fromNelson [1975] 1995:23)

The new spatial organization of the factory meant that workplacefragmentation reflected the ascent of managerial control at the ex-pense of skilled workers’ control. Before 1900, work revolved aroundthe skilled craftsman; after the turn of the century, it was the workerswho revolved around the production process (Gartman 1993; Meyer1982). In the auto industry, even before the advent of the assemblyline, Henry Ford had reorganized the shop floor in a “progressivelayout” system (Gartman 1993:33). In the machine-tool shops, oper-ators were confined to a single machine, and the machines were bunchedtogether. “Unnecessary” movement was eliminated, as operators werelimited to a single workstation, which could include two or more of thesame type of machine. As a result, direct supervision became mucheasier. Most significantly, “progressive layout … ensur[ed] a rapid,continuous flow of work past operators.” As such, it was a crucialprecondition for assembly-line production (Gartman 1993:33). Ofcourse, the reorganization of work in the auto industry cut both ways.The power of the skilled worker was undermined, but a large andincreasingly militant stratum of semiskilled workers was created.Thanks to the shop-floor geography of continuous-flow production,this stratum would assert a decisive leadership role in future classstruggles (Edwards 1979:128).

Beginning in the 1880s, industrial capitalists faced a crisis of controlon the shop floor, brought about by the rapidly growing output ofgiant factories, concentrating unprecedented numbers of workers under


the same roof. The scale of production had grown enormously, but theorganization of production had not. The artisanal mode of organiza-tion, which pivoted on the technical expertise and coordinating func-tion of skilled workers, proved a formidable barrier to industrialists’aggressive efforts to expand production and extract surplus value. Thegrowing concentration of workers, combined with attacks on skilledworkers’ control over the labor process and concomitant efforts toincrease productivity through the speed-up, proved to be favorableterrain for labor unrest. Capital responded to the deepening controlcrisis with strategies aimed at reorganizing the relations betweenworkers in the production process not only by separating conceptionfrom execution as much as possible, but equally by introducing a newspatial form, the giant factory, which made possible new shop floorlayouts that favored capitalists’ rather than workers’ control.

As we shall see, these transformations of the production processwere, in turn, made possible by broader changes in regional industrialgeography.

Moving Capital and Staying Put: The ChangingGeography of Industrializing AmericaAntebellum factories were usually run by owner-operators. Thatchanged after the Civil War. A new industrial bourgeoisie emerged,whose power derived from its command over increasingly large-scaleindustry. While industrial capital’s power was rooted in particularlocales of industrial production, industrial capitalists were no longerbound to these locales. Industrial capital’s unique power in this eraflowed from its ability to command industrial space from a distance.This was made possible by the growing concentration of capital, a re-newed communications–transportation revolution, and a new relation-ship between town and country on a national and world scale (Mumford1938). These developments allowed capital to move production facil-ities “with strategic regard for working-class power, while capitalistclass members relocated to centralized urban environs, stabilized theirliving arrangements, and began the long-term process of building classinstitutions” (Lembcke 1995:158).

Hence, one important consequence of the separation of ownershipfrom direct control was that capitalists could stay put even as theircapital became more mobile:

The separation of labor and capital allows for the independentunification of units of labor and the independent unification of unitsof capital, within the limits posed by the properties inherent in eachelement. Because the value of capital can be converted to a moneyform (liquefied) it can be merged, whereas labor, being a humanelement, can only be unified (collectivized) through social organization,

184 Antipode

a process that is more problematic than the merging of capital.Moreover, the intervention of capital in the processes of collectiv-ization disrupts and retards the formation of the working class.Those interventions are intended to stall the progression of working-class collectivity and the history of those interventions constitutesmuch of the story of US capital–labor relations. (Lembcke 1995:158)

The growing stability of capitalists relative to capital enhancedbourgeois class power, insofar as all social classes rely on social net-works for the assertion and defense of class interests. Up to the late 19thcentury, all classes’ social networks were primarily local.5 With thegrowing concentration of capital, the separation of ownership fromcontrol, and new transportation and communications technologies,this changed for industrial capitalists. No longer obligated to livenear their factories, industrialists could move to big cities in greaternumbers and begin to develop more cohesive intraclass networks anda class consciousness that transcended local conditions. “Thus, whenfamilies that owned steel mills around Pennsylvania merged theirassets, they moved to Philadelphia, where they have remained for ahundred years. Capitalist class members, of course, move around. Buttheir wealth … enables them to diminish the effect of distance onsocial relationships; eg, costly travel and communication networks areno object” (Lembcke 1995:161; see also Lembcke 1991–92:433).

While capitalists were settling in the major cities, those major citiesthemselves were undergoing a major transformation. The late 19thcentury was the era of the industrial city (Gordon 1978).6 In contrast tothe commercial city of the antebellum period, the industrial city cor-responded to the demands of mass production capitalism in two mainways. First, it offered proximity to consumer markets and therebyfacilitated large-scale production. Second, it offered a relatively largereserve army of labor (Gordon 1978:37). While manufacturing em-ployment increased as fast or faster in midsized relative to large urbancenters until the end of the Civil War, by 1870 industrial productionwas concentrating in a few major cities such as Chicago, Cleveland,and New York. Between 1860 and 1900, the ten largest industrialareas’ share of national value-added in manufacturing increased from25% to 40% (Gordon 1978:39). While new power sources such assteam and electricity were important in concentrating industry in bigcities (Laurie 1989:115), the growing industrial dominance of thelargest cities was rooted in the need to resolve the looming “controlcrisis” discussed in the previous section.

Industrial capitalists concentrated manufacturing capacity in the bigcities because large urban space enhanced their ability to extract sur-plus value relative to medium-sized towns, where workers enjoyedmore community support during labor disputes. In the smaller cities,


industrial capitalists faced a number of obstacles not present in thelarger ones. Ideologically, they were challenged by preindustrial socialclasses whose worldviews were markedly different from those of theindustrialists; middle strata in these towns often supported workersagainst industrial capital (Gutman 1976:233–292; Laurie 1989). Inthe bigger cities, industrial capital had more success in winning overmiddle strata, whose fortunes were more closely tied to the newcapitalist order. The reserve army of labor tended to be larger andethnic divisions within the working class stronger (Kolko 1984). Con-sequently, the balance of class power in the big cities initially favoredcapital over labor. Capitalists in the larger cities “suffer[ed] fewerlosses during strikes, achieve[d] greater discipline over their regularfactory, and in general, extract[ed] more surplus value from theirworkers” (Gordon 1978:41).

Yet the industrial city generated some fatal contradictions. Hugenumbers of industrial workers were concentrated in ever-larger pro-duction units. These larger production units were concentrated ingrowing cities, whose share of US population tripled between 1850 and1910, accelerating sharply in the last two decades of that period(North, Anderson and Hill 1983:145; Walton 1996:102). In this respect,the US differed little from other major industrializers such as Russiaand Germany, where growing numbers of workers were concentratedin large factories (Barraclough 1967:51–52; Wolf 1969:75).7 Thenumber of strikers in the US more than tripled between 1881–1885and 1901–1905, and the geographical spread of strike activity nowincluded the Midwest as well as the Northeast (Earle 1992:422–428;Gordon 1978:46–47).

By the turn of the century, rising labor unrest had led capitalists torelocate their factories to industrial suburbs—often creating them asa result. Between 1899 and 1909, central city manufacturing employ-ment increased by 40.8%, while ring employment rose by 97.7%(Gordon 1978:46–47). “Once installed at a sufficient distance fromthe center of labor agitation, these firms achieved a measure of in-sulation from the epidemics of central-city strike activity to which theyhad previously been vulnerable. In moving to the suburbs, corpor-ations sought to isolate their work forces from central-city unrest”(Gordon, Edwards and Reich 1982:138–139). At first, employers experi-mented with company towns, but the 1894 Pullman strike souredmany capitalists on this option (Ashton 1978:71). Increasingly, indus-trial capitalists moved their plants to production sites just outside bigcity limits. Between 1899 and 1905, “new suburban manufacturingtowns were being built in open space like movie sets. Gary, Indiana,constructed from 1905 to 1908, is the best known example” (Gordon1978:48). In these industrial suburbs, capital could “reassert botheconomic and political control” (Ashton 1978:71).

186 Antipode

The crucial precondition for this move from the cities to the suburbswas the growing concentration of capital, especially the 1898–1902merger wave (Edwards 1979:44–45). While the concentration of cap-ital was ongoing throughout the industrial capitalist world, it wasmost advanced in the US (Blackford 1988:55, 72, 80; Hobsbawm 1987:43–44; Schmitz 1993:46–48). The merger wave put far more capitalinto the hands of capitalists, thereby increasing their capacity to moveproduction sites away from high class-struggle areas (Gordon1978:50). Indeed, industrial consolidation and class struggle frombelow were but two sides of the same coin. “The list of strikes [aroundthe turn of the century] reads like a roster of consolidations: the rail-roads, McCormick, Carnegie (Homestead), Pullman, General Electric,US Steel, International Harvester” (Edwards 1979:50). Relocationdovetailed with a renewed employers’ offensive, taking the form of avigorous open-shop drive by manufacturers’ organizations (Montgomery1979:26, 57–63; Nelson 1995:129–130).8 Not coincidentally, the mer-ger wave continued to favor increased plant size in the early 20thcentury (Gordon, Edwards and Reich 1982:133).

The connection between capitalist relocation to industrial suburbsand workers’ struggles provides us with a good opportunity to stepback for a moment to ask how we might conceptualize the relationshipbetween transformations of the technical and social divisions of labor(Marx 1977:470–480; see also Marx and Engels 1970). How are trans-formations of society and transformations of the labor process dia-lectically bound? The literature on labor process transformation viewsskill reduction as the historical process of class conflict through whichcapitalists seek to wrest a progressively greater amount of controlfrom workers and in so doing increase the rate of exploitation(Braverman 1974; Edwards 1979; Marglin 1974). Missing from thisliterature is a consideration of the geographical dimensions of classstruggle, in particular the distinctive geographical bases of class power.Deskilling is a crucial capitalist modality of spatial control and power.When combined with the concentration of capital and the separationof residential ownership from direct control, deskilling meant in-dustrial capital could draw upon a much larger pool of workers overmuch broader geographic space. Workers could now be drawn fromthe far corners of the globe, and found their way to the US in growingnumbers during this era. These new workers, many lacking industrialjob experience, were suddenly employable in huge numbers. By theturn of the century “almost anyone could work in the large factoriesbecause jobs in those factories increasingly required only nominalskills” (Gordon, Edwards and Reich 1982:118). This also meant thatemployers could use ethnic hiring policies to set workers against eachother (Kolko 1984). A larger reserve army tended to keep wages fromrising too fast, although real wages in the advanced capitalist centers


continued to increase throughout the period (Hobsbawm 1987),especially in the US relative to Europe (Montgomery 1987:172). Evenin 1900, real wages in the US were 65% higher than in Britain, andBritain’s wages were relatively higher than those on the Continent(Stalker 1999:12).

The combined result of residential stability for the bourgeoisie, skillreduction for workers, and the separation of ownership from controlwas that industrial capital could now move productive capital as astrategic weapon in the class struggle. This had two big impacts on theclass struggle. First, industrial capital could “jump scale”—circum-venting and, to a certain extent, dismantling “historically entrenchedforms of territorial organization” (Brenner 1999:62). In this case,jumping scale meant that industrialists developed social networks andinstitutional forms that allowed them to operate on a regional andeven national scale much more effectively than could workers’ organ-izations (Wilentz 1984:16). By the turn of the century, and especiallyafter World War I, capital was able to subject the labor movement to“an awesome attack by the employers, the courts, and the states,” notto mention the ideological assault of so-called welfare capitalism(Green 1980:101), effectively rolling back the gains of the previoustwo decades. Crudely put, “capital was more broadly organized thanthe workers, and the workers lost” (Edwards 1979:50).

The second decisive effect of industrial capital’s mobility was theconcentration of the industrial bourgeoisie in the big cities and thepower it could wield over industrial, working-class suburbs. Residen-tial stability was an important geographical aspect of industrial cap-italist class formation. The flip side of this process was that industrialcapital could now disrupt working-class communities by relocatingfactories, thereby disrupting working-class formation. It is, of course,true that capital could not then—and to this day cannot—relocate atwill. Recall, however, that this was a period of rapid industrial ex-pansion. The old modes of industrial organization were becomingincreasingly costly, owing to rising working-class unrest and escalatinginterenterprise competition on an increasingly national scale. Thus,during the era of transition to monopoly capitalism, the costs of aban-doning established centers of production were reduced considerably.

Capital’s ability to relocate in this era was so important becauseworking-class consciousness develops over time in particular places.“In short, it is the length of time a group of working people reside in aparticular place … that is the essential combination of factors in theformation of social solidarity” (Lembcke 1995:163, emphasis added;see also Storper and Walker 1989:157). Lembcke (1991–1992, 1995)estimates this length of time to be two to three generations, aboutfifty years, because consciousness develops over time through theintergenerational transmission of values in specific places. Because

188 Antipode

working-class families have relatively fewer resources to overcomedistance, this intergenerational transmission must occur locally. Thedisruption of local ties through capital relocation therefore disruptstransmission, which consequently stunts the development of militantworking class politics. For capital, the fact that workers’ consciousnessdevelops in specific locales makes “external” solutions through relocationrather than “internal” solutions of restructuring the preferred strategy:

Internal solutions are inherently problematic [for capital]. The slateof local history can never be wiped entirely clean … Therefore, in-dustries frequently opt for radical locational shifts to places wherethey can create new employment relations and labor politics. In thisexternal solution, new communities of firms and workers are con-structed on fresh terrain, usually in regions that have not hithertohad an important presence in the industrial economy. Workers mustin effect chase after these industries and build territorial solutionsanew. (Storper and Walker 1989:179)

From this perspective we can make sense of the connection thatLembcke draws between working-class militancy in a particular placewith roughly fifty-year cycles of industrial relocation (1995:163–164).This seems about right. We have already seen how industrial capitalmoved from the big cities to industrial suburbs in the early 20thcentury. Capital again began to move after World War II, with roughlysimilar disorganizing effects on the working class (Davis 1986:117–121).What remains to be seen today is whether Lembcke’s fifty-year cycleis subjected to the same kind of “time–space compression” that hascharacterized other capitalist cycles (Harvey 1989), and if so, whetherworkers or capitalists fare better because of it.

By the early 20th century, industrial capitalists had dramatically in-creased their mobility relative to workers. The growing concentrationof capital, combined with the rapidly developing transportation andcommunications infrastructure, made possible the separation ofownership from direct control. The bourgeoisie was now free to con-gregate in big cities, which had a positive impact on their social coher-ence and consciousness as a regional—even national—class. Thesedevelopments also allowed capital to move production from relativelyhigh class-struggle areas to relatively low class-struggle areas, firstfrom the medium-sized cities to the big cities, then from the big citiesto the new industrial suburbs. This spatial strategy had a profoundlydisorganizing impact on workers, whose power rested on place-bounded social networks.

The reorganization of the labor process and the movement of factoriesfrom the cities to the suburbs at once rested upon and made possiblemassive immigration. The sources of the agrarian transformationsthat allowed for the upsurge in immigration can be found in the


changing division of labor between town and country in the US, andin the world-economy as it entered a new era of global expansion andrestructuring.

Remaking the US Town–Country Division of Labor in World PerspectiveTransformations of the labor process and the geography of industrialregions in the US were local expressions of a profound shift in theworld capitalist system9—a shift that was part cause and partconsequence of US capitalism’s rise to world power. In this section, Iexplain how the reconfiguration of the town–country division of labor,within the US and on a world scale, at once shaped local and regionaltransformations and rendered possible US capitalism’s ascent. Twoaspects of this reconfiguration of the rural–urban dialectic are high-lighted: (1) the expansion of scale, principally in the form of US ex-pansionism in the American West; and (2) the production of ainterregional division of labor, marked especially by the rise of theMidwest as the epicenter of agroindustrialization as the distinctivefeature of US capitalist development.

The Emergence of an American Empire: “Manifest Destiny” as Global ExpansionThe role of US territorial and capitalist global expansion in shapingthe world division of labor was not confined to the period after theonset of the Spanish–American War in 1898. Over the previous threedecades, US governmental and business agencies had set aboutcreating a dynamic and cohesive national political economy throughthe subordination of the American South and, most importantly,through the conquest and colonization of the American West. Thiswas the precondition for the projection of US power overseas.

In contrast to the European powers, as well as ascendant Japan, 19th-century American expansion took the unusual form10 of national ratherthan colonial expansion. This gave American capital tremendous advantages. National-continental expansion was considerably cheaperthan overseas colonial expansion. More significantly, the creation of a continental economy—unprecedented in modern world history—gave the US a distinct competitive advantage in the struggle for worldsupremacy that attended the crisis of British hegemony in the later 19thcentury:

The US state enclosed an economic space that was not only muchlarger and more diversified, but also far more malleable than theeconomic space enclosed by Imperial Germany—a space, that is,that could be depopulated and repopulated to suit the requirementsof high-tech agricultural production more easily than the smaller andmore densely populated German economic space could. In the course

190 Antipode

of the [1873–1896] Great Depression, this competitive advantage trans-lated into the progressive displacement of German by US agriculturalsurpluses so that the already larger US domestic market grew muchfaster than Germany’s. (Arrighi 1994:292)

In other words, the US was in the right place at the right time—the eraof the “second industrial revolution,” marked by the development ofmass production systems based on the new technologies of steel,electricity, the internal combustion engine, and so forth. These newproduction systems demanded a new relationship between core andperiphery. “It was no longer a question of exchanging Europeanmanufactures—predominantly textiles—for traditional oriental andtropical products … Industry now went out into the world in search ofthe basic materials without which, in its new forms, it could not exist”(Barraclough 1967:54). This was a decisive shift from the “first” indus-trial revolution. “Whereas in the eighteenth century Britain providedall the essential raw materials for its industries … in 1913 Britain wasself-sufficient only in the supply of coal” (Woodruff 1973:663). Theascendant steel, chemical, electrical, and automobile industries pro-vided a decisive impetus to imperial expansion in order to secure rawmaterial supplies as well as markets.

What set the US apart from the era’s other leading capitalist powerswas not imperialism or lack of it, but the existence of sparsely settledyet ecologically rich contiguous space. Thus the American empire wasbuilt initially on the frontier. “Until 1898 American ‘imperialist’ in-vestments had been limited mainly … to the American West” (Cochranand Miller [1942] 1961:312). US imperialism was distinctive but not“exceptional.” Between 1870 and the early 1900s,

the American West was the great natural-resource reservoir and theinvestment arena for eastern US and western European capital. Inthat sense, the region was part of the wider subordination of colonialsectors to the requirements of metropolitan-based economies. With alargely monocultural focus based on the exploitation of natural re-sources for export, those rural, hinterland economies were linked tocenters of industry and manufacturing in a decidedly dependent way.Hence, the great advances in production and in the accumulation ofwealth in the Atlantic-centered industrial economy depended heavilyon resources from the American West … [I]t was the world ofeastern (and European) capital—not the sturdy work of the solitaryprospector or the sodbuster—that provided the major impetus tochange in the West. (Robbins 1994:62–63, emphasis added)

The railroad was the means by which the American West was trans-formed into an arena for massive capital investment, a growing con-sumer market, and a major supplier of agricultural products and rawmaterials. With the elimination of the last effective Indian resistance


in the 1870s and the completion of the first transcontinental railroad in1869, it became possible to create, in effect, a region that correspondedunusually well to the specific needs of mass production capitalism.American railroads “seemed to create new spaces, new regions ofcomprehension and economic value, and finally to incorporate a pre-historic geologic terrain into historical time” (Trachtenberg 1982:58–59).

The region’s size, and the relatively greater mobility of both laborand capital within it—thanks to the railroad—provided US capitalwith a decisive advantage over its international rivals, who were forcedto rely on more costly forms of colonial expansion to achieve the sameends. The geographical distinctiveness of US capitalism is revealed inthe contrast between American and European rail networks. In thefirst place, US railroad expansion reversed the relationship betweenlocomotive and track that prevailed in Britain. Americans innovatedby laying track as cheaply as possible—at as little as one-tenth the costof British railroads—and by building stronger locomotives capable ofbalancing on such cheap track (Hugill 1993:174–175). Whereas Euro-pean “mechanized transport appropriated existing roads and horsetracks as it overturned an older society and culture, [in the US] therailroad seemed to ‘open’ places for settlement, for raw materials andtransport to markets” (Trachtenberg 1982:58). It was no accident thatthis American mode of railroad expansion was subsequently adoptedby British capital overseas (Vance 1988:129).

The US railroad network—which by 1915 was nearly double thecombined mileage of the six other leading national systems (Headrick1988:55)—made possible the massive extraction of natural resourcesfrom the American West crucial to US industrialization: iron, copper,zinc, gold, silver, coal, and (above all?) oil (Agnew 1987:49). “Theribbons of steel that everywhere penetrated the mountain West …[permitted] the rapid expansion of an integrated, metropolitan-industrial system surrounded by peripheral population centers inresource-rich regions that paid tribute to the manufacturing core”(Robbins 1994:88; see also Agnew 1987:111). From this perspective,throughout the American West it was not so much “modes of pro-duction” but “modes of extraction” (Bunker 1985:chapter 2) thatdetermined the region’s capitalist formation (White 1991:243).

Railroad expansion provided, in Alfred Chandler’s phrase, “the firstmodern business enterprises” (1977:chapter 3). As such, it wouldestablish “important precedents” for the large industrial firms thatwould emerge in the later 19th century. Beyond this, railroads tendedto annihilate regional markets in favor of national—even international—markets, at once intensifying interenterprise competition andforcing industrial firms to develop new means of coordinating andadministering far-flung production and distribution activities. The geo-graphical expansion and integration of the nationwide social division of

192 Antipode

labor also shaped the technical division of labor: “Expanding markets… encouraged the use of more complex machinery in manufacturing… This new and increasingly complicated machinery [—embodying‘complicated’ knowledge that once resided in the heads of workers—]in turn spurred further increases in output and so provided anotherpressure for expansion and continued growth” (Chandler 1962:23).

These pressures culminated in a new phase of global expansion oncethe US emerged from depression in 1896. US-based transnationalfirms held over $1 billion in foreign assets by 1895, and $2.5 billion by1914; this latter figure amounted to some 7% of the US gross nationalproduct, a level that would not again be reached until 1966 (Barff1995:53; Chandler 1977:369; Zinn 1980:293). As we might expect,given the interplay between the competitive pressures arising fromvarious layers of the social and technical divisions of labor just dis-cussed, the largest and most organizationally advanced firms were atthe vanguard of overseas expansion. By 1902, “Europeans were al-ready speaking of an ‘American invasion’” (Arrighi and Silver 1999:130).

The Centrality of Agrarian TransformationIn the postbellum era, American expansionism shaped and was shapedby the capitalist transformation of US agriculture and the emergenceof a massive agroindustrial complex in the Midwest. American cap-italist agrarian transformation, mediated by British free-tradepolicies, began to affect agrarian societies worldwide, especially inEurope. American grain entered the world market in the later 19thcentury, precipitating a prolonged crisis in European agriculture. Theresulting social disruption drove millions of peasants from the land.Many of these found their way to North America, providing the laborpower necessary for the mass production revolution.

The development of American capitalism, then, was distinctive inseveral respects. In the first place, in contrast to the British “agri-cultural revolution” of the 17th and 18th centuries (Thompson 1968),agricultural innovation directly fueled industrialization through amassive farm implements industry (Post 1995; Pudup 1987). Americanindustrialization’s historical specificity is found in its agroindustrialcharacter. Its geographical specificity is found in the emergence of theMidwest as a new region. Secondly, while American agriculture gen-erated a modest surplus population (Laurie 1989), the main source oflabor power for agroindustrialization was found in those agrariansocieties rendered vulnerable to the leveling effects of a world grainmarket. Finally, the emergence of the Midwest as the agroindustrialpowerhouse of the US—and, eventually, the world-system—hingedon the conquest of the American West, which supplied the necessaryraw materials and gave the US the decisive edge in the competitionfor world leadership in the 20th century.


In contrast to its rivals, the US became at once a major industrialpower and a major agricultural exporter. The deepening commercial-ization of agriculture pushed forward a new capitalist division of laborwithin the national economy that allowed for the transition to mass-production monopoly capitalism. “So long as the bulk of thepopulation lived on farms or in small towns, commodity productionconfronted a barrier that limited its expansion” (Braverman 1974:272). Among other things, rising agricultural productivity permitted arapid reallocation of the country’s labor force away from agriculture.While agriculture occupied just over half the US workforce in 1870, itsshare had fallen to less than a third by 1910 (White 1991:244).

Thus, capitalist agrarian transformation forms a decisive moment inthe creation of a “universal market” (Braverman 1974:271–283). Therapid growth of food processing, outstripping agricultural expansion,was a decisive factor in Midwestern industrialization by midcentury,and meatpacking’s “disassembly line” was in motion by the 1830s(Cronon 1991:211, 229, 243; Giedion 1969:93–95; Page and Walker1991:292–293). The social basis for such agroindustrialization hadbeen established even earlier, during the agrarian class struggles of the1780s and 1790s. This was the crucial moment of primitive accumu-lation in the US (Post 1995), dramatized by the defeat of farmers’revolts against taxation—and therefore the necessity of market pro-duction—such as the Whiskey Rebellion. With the victory of capitalistinterests over subsistence-oriented freeholding agriculture in theNorth, by the 1840s and 1850s Northern and Midwestern farmers hadbecome increasingly dependent on commodity production rather thandiversified home production (Friedmann 1978; Post 1995). Thiscommercial orientation laid the groundwork for the massive railroadconstruction that characterized mid-19th-century America. Railroadstransported crops to urban markets—and by steamship to increasinglydistant markets in Europe (Wolf 1982)—and carried back from thecities consumer goods such as clothing and producer goods such asthreshers and reapers. As in industry, by creating more integratednational and international markets, the railroads intensified thecompetitive pressures faced by American farmers, and with thesepressures came new insecurities.

After the Civil War, millions of settlers flowed into the West.Between 1860 and 1900, the number of farms grew from 2 to 5.7million, and cultivated land increased from 407 to 838.5 million acres (Shannon, 1968:51). The mass market—the foundation of themass production system that necessitated scientific management—was,therefore, only partly urban. Even in 1910 a majority of the country’spopulation still lived in rural areas, although that would change byWorld War I (Walton 1996:102). The mass market was still largely a country store (Page and Walker 1991:309). As farm households

194 Antipode

became more specialized and dependent on heavy capital inputs suchas reapers and threshers, they began to consume an ever-largeramount of producer goods and, because time was increasingly scarce,they consumed progressively more consumer goods (Friedmann 1978).

As farm households in the North and Midwest became increasinglyspecialized commodity producers, interregional inequality rosesharply. All regions saw some industrialization, measured by the num-ber of manufacturing workers. Yet there was tremendous disparity,especially between the Midwest and Northeast on the one hand andthe primarily extractive and agricultural economies of the South andfar West on the other. While one-third to one-half the workforce inthe former regions was employed in manufacturing by 1910, just one-eighth to one-quarter of the latter regions’ employment was indus-trial. In 1910, the Northern industrial regions contained 48% of USpopulation but were responsible for 68% of manufacturing output.The South fared particularly poorly; it was “drained of its resources”and thereby “sank into a tributary condition as the most backwardsection of the national economy.” Despite westward expansion, theSouth’s share of national primary-sector employment continued torise in the later 19th century (Agnew 1987:110–111). Moreover, the“export of Southern cotton underwrote northern economic expan-sion,” thanks to tariff barriers that raised prices for industrial goods(Agnew 1987:114).

If the South helped pay for industrialization, and westward ex-pansion helped consolidate the new continental division of labor, itwas in the Midwest that US capitalists pioneered the distinctive pat-tern of late 19th-century American capitalist development, agroindus-trialization. Building upon the increasing market orientation andagricultural innovation of its farmers between the 1830s and 1850s,the Midwest emerged as a dynamic agroindustrial complex based onthe complementarity of specialized commercial agriculture and urbanindustrialization centered on both producer- and consumer-goodssectors, especially food-processing and the manufacture of agricul-tural implements (Page and Walker 1991; Post 1995; Pudup 1987).Thus, Midwestern agroindustrialization came to embody the mostbasic and well-known characteristics of US capitalism as it began itsrise to world hegemony: (1) a pattern of “autocentric” developmentwhereby capitalist development proceeded according to the nationalintegration of agriculture and industry, in contrast to the extroversionof British capitalism, for which free trade was essential for its survival;and (2), following the consolidation of this agroindustrial complex,the emergence of “Fordism,” that system of mass production that socaptured the attention of contemporary European observers (Arrighiand Silver 1999; Friedmann and McMichael 1989; Gramsci 1971:277–318; Page and Walker 1991).


Agrarian transformation, which constituted the basis of US indus-trial prowess, was itself part of a global movement towards capitalistagriculture that took place under British hegemony. Between the1850s and 1880s, world food exports rose from 4 million tons a year to 18 million, and to 40 million tons in 1914, a level that would notagain be attained until the 1950s (Ponting 1991:244–245). In theafter-math of Britain’s repeal of the Corn Laws (1846), the worldgrain market experienced rapid growth, much of it supplied byproducers located in “frontier” zones. Russia and the US, the world’sleading wheat producers, exported between 8 and 20 times as muchwheat in the early 1870s as they did in the 1840s, and world wheatexports would increase another sixfold between 1873 and 1929(Friedmann 1978:546; Hobsbawm 1975:191; Wolf 1982:319). Americanwheat farmers exported fully one-third of their crop by 1900 (ASHP1992:52).

Especially in grain-producing regions, rapid world-market growthentailed an acceleration of social differentiation and insecurity in theUS and Europe. Indeed, American developments were functionallyrelated to subsequent European agrarian crises. In the US, farmersinvested in machinery (primarily reapers) and sought to acquire newland (even at inflated prices), resulting in further indebtedness andinsecurity after the boom had subsided (Post 1997).11 The mechan-ization of agriculture led to sharply rising productivity. “Using auto-matic binders to cut and tie bundles of wheat, the … farmer couldplant 135 acres of wheat in 1890, nearly 20 times more than in 1860”(ASHP 1992:52; see also Friedmann 1978; Laurie 1989:122–127).Higher productivity in turn led to a crisis of overproduction on both aworld and a national scale by the 1870s. Prices fell sharply, which wasvery bad for farmers but good for urban working-class consumers, andvery good for the industrial capitalists who paid their wages (Agnew1987:52–54; Friedmann and McMichael 1989).

In a situation where grain producers through the world were inte-grated into the same competitive market, US agriculture’s product-ivity quickly set in motion agrarian upheaval in Europe once theBritish-led boom turned to depression in the 1870s. By the 1870s and1880s, “American wheat, sold in Europe at lower prices than thedomestic product, brought on a crisis in European peasant agricul-ture, sending a migrant stream of ruined peasants to seek new sourcesof livelihood in the burgeoning Americas” (Wolf 1982:319; see alsoAgnew 1987:53; Woodruff 1973:704). The crisis was particularlysevere in Eastern and Central Europe (Potts 1990:132). Between 1880and 1920, 23.5 million immigrants arrived in the US (Agnew 1987:52).So great was immigration from southeastern Europe that the latterhad “become part of the American industrial economy” by the turn ofthe century (Montgomery 1980:210).

196 Antipode

These immigrant workers provided the labor power upon which the20th century’s leading industries were built. In so doing, they re-inforced the processes of agroindustrial specialization and rational-ization that proved so disruptive to the older “producing classes”—especially skilled craft workers and farmers. The emergence of anew working class shaped by the new basic industries, the risingdiscontent of the old working class, and the continual disruptions toworking-class consciousness brought about by industrial capital’s newmobility all contributed to a new round of bitter class struggles thatshaped class structure and conflict in the 20th century.

ConclusionRather than focus on one dimension of social change confined to asingle geographical scale, I have instead offered an alternativeaccount of American capitalism between 1865 and 1920 that showshow social transformations at one scale were inconceivable withoutsimultaneous transformations at other scales. Scientific managementwas necessary to solve the “control crisis” of the 1870s and 1880s,but the “solution” to the crisis involved far more than reorganizingand rationalizing the social relations of production narrowly con-ceived; it required new factory forms, which in turn were madepossible by the new technologies of the “second industrial revolution,”such as electricity and steel. Yet, new factory forms were useless ifindustrial concentration brought together too many workers too fast,leading to costly labor unrest, so relocation from the cities to indus-trial suburbs became a crucial element in capitalist restructuring.These new production centers could exist, however, only withdependable supplies of raw materials, and only with dependablemarkets. This required a new, national-scale division of labor betweentown and country that, for the first time, integrated the continentalexpanse of the United States. Natural resources were extracted fromthe West. Consumer- and producer-goods markets were created as aresult of the proliferation of small-farmer commodity productionthroughout the country, especially in the Midwest. As they wereintegrated into increasingly competitive (and global) markets, theseproducers were forced to become ever more specialized and, as aconsequence, ever more capital-intensive. This promoted industrial-ization in the crucial capital-goods sectors, because farmers nowneeded threshers, reapers, binders, and other implements to survive ina highly competitive market. Specialization also meant that farmhouseholds now needed to purchase many more items for householdreproduction (including many foodstuffs) that they had previouslymanufactured at home or traded for locally. Moreover, agriculturalspecialization produced foodstuffs relatively cheaply, allowing for amassive expansion of the industrial working class.


All these developments shaped and were shaped by developmentsin the world capitalist system. The transformation of American agri-culture, in particular, had both causes and consequences that werefound beyond the boundaries of the US. Britain’s repeal of the CornLaws was an important factor in creating a world agricultural market,which expanded dramatically over the next seventy years. US grainexports rose dramatically over the following 15 years (North 1966:284). “By 1870 the American economy depended so much uponforeign markets for the agricultural surplus that the ups and downsfor the next thirty years can be traced in large measure to the successor failure of marketing each year’s wheat and cotton crop. No matterhow many markets could be found, more always seemed to beneeded” (LaFeber 1963:9–10). Beyond the US, American agriculture,compelled towards rising productivity by its integration into nationaland world capitalist markets, precipitated major agrarian crises inEurope. Far from undermining US development, such crises onlyserved to reinforce it further, as European immigrants provided areadily exploitable labor supply for rapid industrialization.

All of this social change “from above” encountered the resistance ofworkers and farmers, whose struggles unfolded differentially accord-ing to their position in the labor process, the relative fixity or mobilityof capital, the power of the state, their organizations’ capacity to forgea unifying culture, and the spatial concentration of workers. In thecourse of postbellum capitalist development, regional movements ofworkers and farmers dissolved in favor of organizations increasinglynational in scope, the consequence of a new transportation and com-munications infrastructure and the integration of the national econ-omy. Industrial capital clearly had the upper hand, exercising its superiormobility to disrupt working-class consciousness and organization. Yetthe very strategies capital pursued to undermine labor’s challengeswould, in a matter of decades, confront capital with successive wavesof working-class revolts, first in the 1930s, then in the later 1960s andearly 1970s, and, quite possibly, again in the early 21st century.

AcknowledgmentsI am greatly indebted to Diana Carol Moore Gildea, with whom aninitial draft of this paper was written, for comments on successivedrafts. I would also like to thank Giovanni Arrighi, Ben Brewer,Carolyn Cartier, Brett Clark, John Bellamy Foster, Peter Grimes,Andrew Herod, Ho-fung Hung, Yukio Kawano, Dan Pope, BeverlySilver, and an anonymous reviewer for Antipode. Special thanks go toDick Walker for a careful reading of this paper in draft.

Endnotes1 This is not to say that there was no uneven development within the US, as we shall see.

198 Antipode

2 This paper’s angle of vision prioritizes the town–country division of labor as a dialecticof sociospatial change distinct from, but mutually relational with, the technical divisionof labor (following Marx and Engels, especially 1970; Marx 1977:470–480; see alsoMoore 2000:125–129). Unfortunately, Marx and Engels specify neither the scale atwhich the former dialectic operates nor the factors that distinguish “town” and“country” from each other. As a preliminary hypothesis, I would argue that the town–country dialectic operates at multiple scales, from city–hinterland relations at theregional scale to national formations to the world-economy. Furthermore, I suggestthat O’Connor’s (1998) distinction between conditions and forces of production isuseful for distinguishing town and country. In the cities, land is primarily a conditionof production (eg land ownership is a condition of industrial production). In thecountryside, land is not only a condition but also (primarily) a force of production; it isan instrument one can use for primary production (eg, agriculture).3 “Taylorism” here is but a convenient term to reference the broad range of ideas andpractices corresponding to the emergence of a highly rationalized and abstractmanagerial approach to production. I make no claim that Taylor’s ideas were adoptedvery widely.4 In 1869, “almost half the American manufacturing establishments drew power fromwaterwheels and turbines” (Trachtenberg 1982:57).5 This was true with the exception of finance and (some) mercantile capitalists, whosenetworks had been transnational from the Middle Ages (Arrighi 1994).6 For Gordon (1978), urban development is discontinuous and determined by classstruggles over the labor process rather than by a reified logic of agglomeration ortechnological change. In other words, urban development and uneven developmentbetween the country and the city stem from forces endogenous to the class struggle and capital accumulation. Gordon (1978) identifies three main stages of capitalaccumulation that correspond to three stages of US urban development: the commercialcity, the industrial city, and the corporate city. 7 While all three countries generated powerful working-class movements, US capitaldid not have to contend with a serious threat of social revolution, in large part becausethe dangers of industrial concentration were partially offset by two factors. First, oncethe new communications and transportation infrastructure was in place, the contin-ental scale of the US economy afforded industrialists relatively greater freedom torelocate manufacturing facilities to more favorable locales. Capital’s greater capacityto command space was complemented by ethnic cleavages among immigrant workers,many of whom saw their stay in American as temporary (Kolko 1984). 8 For example, the Detroit area was such an attractive location for industrial growth inpart because the Employers’ Association of Detroit had by 1912 “virtually obliteratedthe local labor movement outside the building trades” (Nelson 1995:129; see alsoRubenstein 1992:234–235).9 My approach here borrows heavily from Dale Tomich (1990), who stresses that“local” and “global” transformations are irreducible to and, at the same time, mutuallyformative of each other. In his study of sugar and slavery in 19th-century Martinique,Tomich (1990:7) argues for a world-historical approach to the study of local socialtransformation (including the labor process) that treats “world, national, and localphenomena” as analytically distinct but not empirically discrete:

The concern of this study is not simply to examine the interaction of distinct globaland local “levels.” Rather, it attempts to go beyond the fact of the extroversion ofthe colonial economy and its domination by European capital in order to unite thelocal history of plantation slavery … with the history of world capitalism. Instead ofapproaching world, national, and local dimensions as discrete empirical entities, ittreats them as mutually formative parts of a larger whole. Such a focus on the globalensemble of exchange relations, material and social relations of production, and thepolitical mediations between them as a unified, structured, continually evolving


totality permits the historical specification of each of the relations and processesconstituting colonial slavery … The history of Martinique can thereby be under-stood not merely as “local particularism” but as part of the global processes ofcapitalist development. This approach reveals the world historical character of localprocesses while giving specific historical content to the concept of world-economythrough the concrete analysis of particular phenomena. (Tomich 1990:6; emphasisin original)

10 This was true except for Russia, for whom frontier expansion served a broadlysimilar function (see Wallerstein 1989:chapter 4).11 Whatever short-run gains farmers may have earned, the long-run consequence forthem as a class was increased debt and instability (see Luxemburg 1972:410; Shannon1957:9–10). Some five to eight million people left the farms for the cities between 1860and 1890 (Laurie 1989:124).

ReferencesAgnew J (1987) The United States in the World-economy. Cambridge, UK: Cambridge

University PressAronowitz S (1973) False Promises: The Shaping of American Working-Class

Consciousness. New York: McGraw HillArrighi G (1994) The Long Twentieth Century. New York: VersoArrighi G (1999) The global market. The Journal of World-Systems Research

5(2):217–251Arrighi G and Silver B J (1999) Chaos and Governance in the Modern World System.

Minneapolis: University of Minnesota PressAmerican Social History Project (ASHP) (1992) Who Built America? Working People

and the Nation’s Economy, Politics, Culture, and Society. Vol 2. New York: PantheonBooks

Ashton P J (1978) The political economy of suburban development. In W K Tabb andL Sawers (eds) Marxism and the Metropolis (pp 64–89). New York: Oxford Uni-versity Press

Bairoch P (1982) International industrialization levels from 1750 to 1980. Journal ofEuropean Economy History 11:269–333

Baran P and Sweezy P (1966) Monopoly Capital. New York: Monthly Review PressBarff R (1995) Multinational corporations and the new international division of labor.

In R J Johnston, P J Taylor and M Watts (eds) Geographies of Global Change(pp 50–62). Cambridge, MA: Blackwell

Barraclough G (1967) An Introduction to Contemporary History. New York: PenguinBecker W H (1984) 1899–1920: American adjusts to world power. In W H Becker and

S F Wells, Jr (eds) Economics and World Power (pp 173–224). New York: ColumbiaUniversity Press

Blackford M G (1988) The Rise of Modern Big Business in Great Britain, the UnitedStates, and Japan. Chapel Hill: University of North Carolina Press.

Bluestone B and Harrison B (1982) The Deindustrialization of America. New York:Basic Books

Braverman H (1974) Labor and Monopoly Capital. New York: Monthly Review PressBrenner N (1999) Beyond state-centrism? Space, territoriality, and geographical scale

in globalization studies. Theory and Society 28:39–78Brenner R (1977) The origins of capitalism: A critique of neo-Smithian Marxism. New

Left Review 104:25–92Brenner R and Glick M (1991) The regulation approach: Theory and history. New Left

Review 186 (July/Aug.):45–119Bunker S G (1984) Underdeveloping the Amazon. Urbana: University of Illinois Press

200 Antipode

Chandler A (1962) Strategy and Structure. Cambridge, MA: MIT PressChandler A (1977) The Visible Hand: The Managerial Revolution in American Business.

Cambridge, MA: Harvard University PressClawson D (1980) Bureaucracy and the Labor Process: The Transformation of US

Industry 1860–1920. New York: Monthly Review PressCochran T C and Miller W ([1942] 1961) The Age of Enterprise: A Social History of

Industrial America. Revised ed. New York: Harper and RowCronon W (1991) Nature’s Metropolis. New York: W W NortonDavis M (1986) Prisoners of the American Dream. New York: VersoEarle C (1992) Geographical Inquiry and American Historical Problems. Stanford, CA:

Stanford University PressEdwards R (1979) Contested Terrain: The Transformation of the Workplace in the

Twentieth Century. New York: Basic BooksEngels F (1979). The Housing Question. Moscow: Progress PublishersFaulkner H U (1959) Politics, Reform, and Expansion 1890–1900. New York: HarperFriedmann H (1978) World market, state, and family farm: Social bases of household

production in the era of wage labor. Comparative Studies in Society and History20(4):545–586

Friedmann H and McMichael P (1989) Agriculture and the state system: The rise anddemise of national agricultures, 1870 to the present. Sociologia Ruralis 29(2):93–117

Gartman D (1993) The historical roots of the division of labor in the US auto industry.In B Berberoglu (ed) The Labor Process and Control of Labor (pp 21–43). Westport,CT: Praeger

Geyer M. and Bright C (1996) Global violence and national wars in Eurasia andAmerica: The geopolitics of war in the mid-nineteenth century. Comparative Studiesin Society and History 38(4):619–657

Giedion S (1969) Mechanization Takes Command: A Contribution to AnonymousHistory. New York: W W Norton and Co

Gordon D M (1978) Capitalist development and the history of American cities. In W K Tabb and L Sawers (eds) Marxism and the Metropolis (pp 25–63). New York:Oxford University Press

Gordon D M, Edwards R and Reich M (1982) Segmented Work, Divided Workers.Cambridge, UK: Cambridge University Press

Gramsci A (1971) Selections from the Prison Notebooks. Edited by Q Hoare and G N Smith. New York: International Publishers

Green J R (1980) The World of the Worker: Labor in Twentieth-Century America. NewYork: Hill and Wang

Gutman H (1976) Work, Culture, and Society in Industrializing America. New York: VintageHaraven T K and Langenbach R (1978) Amoskeag: Life and Work in an American

Factory City. New York: Pantheon BooksHarvey D (1989) The Condition of Postmodernity. Cambridge, MA: BlackwellHarvey D (1998) The geography of class power. In L Panitch and C Leys (eds) The

Communist Manifesto Now: Socialist Register 1998 (pp 49–74). London: MerlinHarvey D (2000) Spaces of Hope. Berkeley: University of California PressHeadrick D (1988) The Tentacles of Progress: Technology Transfer in the Age of

Imperialism, 1850–1940. New York: Oxford University PressHobsbawm E J (1975) The Age of Capital, 1848–1875. New York: MeridianHobsbawm E J (1987) The Age of Empire, 1875–1914. New York: VintageHounshell D A (1984) From the American System to Mass Production, 1800–1932.

Baltimore: Johns Hopkins University PressHugill P J (1993) World Trade since 1431. Baltimore: Johns Hopkins University PressHymer S (1972) The multinational corporation and the law of uneven development. In

J N Bhagwati (ed) Economics and World Order (pp 113–140). New York: Macmillan


Kinder H and Hilgemann W (1978) Anchor Atlas of World History. Vol 2, From theFrench Revolution to the American Bicentennial. New York: Doubleday

Kolko G (1984) Main Currents in Modern American History. New York: PantheonLaFeber W (1963) The New Empire: An Interpretation of American Expansion

1860–1898. Ithaca, NY: Cornell University PressLaurie B (1989) Artisans into Workers: Labor in Nineteenth-Century America. New

York: Noonday PressLembcke J (1991–1992) Why 50 years? Working-class formation and long economic

cycles. Science & Society 55(4):417–445Lembcke J (1995) Labor history’s “synthesis debate”: Sociological interventions.

Science & Society 59(2):137–173Luxemburg R (1972) The Accumulation of Capital. New York: Monthly Review PressMaddison A (1995) Monitoring the World Economy 1820–1992. Paris: Development

Centre of the Organisation for Economic Cooperation and DevelopmentMarglin S A (1974) What do bosses do? The origins and functions of hierarchy in

capitalist production. Review of Radical Political Economics 6(2):33–60Marx K (1977) Capital, vol 1. New York: VintageMarx K and Engels F (1970) The German Ideology. New York: InternationalMeyer D (1984) Emergence of the American manufacturing belt: An interpretation.

Journal of Historical Geography 9(2):145–174Meyer III S (1982) The Five-Dollar Day: Labor Management and Scientific Control in

the Ford Motor Company, 1908–1922. Albany: State University of New York PressMontgomery D (1979) Workers’ Control in America. Cambridge, UK: Cambridge

University PressMontgomery D (1980) Labor and the Republic in industrializing America: 1860–1920.

Le Mouvement Social 111:201–215Montgomery D (1987) The Fall of the House of Labor: The Workplace, the State, and

American Labor Activism, 1865–1925. Cambridge, UK: Cambridge University PressMoody K (1988) An Injury to All. New York: VersoMoody K (1997) Workers in a Lean World. New York: VersoMoore J W (2000) Environmental crises and the metabolic rift in world-historical

perspective. Organization & Environment 13(2):123–158Mumford L (1938) The Culture of Cities. New York: Harcourt Brace JovanovichNelson D ([1975] 1995) Managers and Workers: The Origins of the New Factory System

in the United States, 1880–1920. 2nd ed. Madison: University of Wisconsin PressNorth D C (1966) The Economic Growth of the United States 1790–1860. New York:

W W NortonNorth D C, Anderson T L and Hill P J (1983) Growth and Welfare in the American Past:

A New Economic History. 3rd ed. Englewood Cliffs, NJ: Prentice-HallO’Connor J (1998) Natural Causes. New York: GuilfordPage B and Walker R (1991) From settlement to Fordism: The agroindustrial

revolution in the American Midwest. Economic Geography 67(4):281–315Parker M and Slaughter J (1988) Choosing Sides: Unions and the Team Concept.

Detroit: Labor NotesPeet R (1987) The geography of class struggle and the relocation of United States

manufacturing industry. In R Peet (ed) International Capitalism and IndustrialRestructuring: A Critical Analysis (pp 40–71). Boston: Allen and Unwin

Pietrykowski B (1995) Fordism at Ford: Spatial decentralization and labor segmen-tation at the Ford Motor Company, 1920–1950. Economic Geography 71:383–401

Pletcher D M (1984) 1861–1898: Economic growth and diplomatic adjustment. In W H Becker and S F Wells Jr (eds) Economics and World Power: An Assessment ofAmerican Diplomacy since 1789 (pp 119–172). New York: Columbia University Press

Ponting C (1991) A Green History of the World. New York: St Martin’s Press

202 Antipode

Post C (1995) The agrarian origins of US capitalism: The transformation of the Northerncountryside before the Civil War. Journal of Peasant Studies 22(3 April):389–445

Post C (1997) The “Agricultural Revolution” in the United States: The developmentof capitalism and the adoption of the reaper in the antebellum US North. Scienceand Society 61(2):216–228

Potts L (1990) The World Labour Market. London: Zed PressPudup M B (1987) From farm to factory: Structuring and location of the US farm

machinery industry. Economic Geography 63(3):203–222Robbins W G (1994) Colony and Empire: The Capitalist Transformation of the American

West. Lawrence: University Press of KansasRobinson R V and Briggs C M (1991) The rise of factories in nineteenth-century

Indianapolis. American Journal of Sociology 97(3):622–656Rubenstein J M (1992) The Changing US Auto Industry: A Geographical Analysis.

London: RoutledgeSayer A and Walker R A (1992) The New Social Economy. Cambridge, MA: BlackwellSchmitz C J (1993) The Growth of Big Business in the United States and Western Europe,

1850–1939. London: MacmillanShannon F A (1957) American Farmers Movements. Princeton, NJ: D Van Nostrand

CompanyShannon F A ([1945] 1968) The Farmers’ Last Frontier: Agriculture, 1860–1897. New

York: Harper and RowShefter M (1986) Trade unions and political machines: The organization and

disorganization of the American working class in the late nineteenth century. In I Katznelson and A Zolberg (eds) Working-Class Formation (pp 197–276).Princeton, NJ: Princeton University Press

Silver B J (1997) Turning points of workers’ militancy in the world automobileindustry, 1930s–1990s. Research in the Sociology of Work 6:43–71

Smith N (2000) “New Globalism, New Urbanism: Uneven Development in the 21stCentury.” Unpublished paper. Department of Anthropology, City University ofNew York

Stalker P (1999) Workers without Frontiers. Boulder, CO: Lynne RiennerStorper M and Walker R (1989) The Capitalist Imperative: Territory, Technology, and

Industrial Growth. Oxford: Basil BlackwellSwyngedouw E (1997) Neither local nor global: “Glocalization” and the politics of

scale. In K Cox (ed) Spaces of Globalization (pp 137–165). New York: GuilfordThompson F M L (1968) The second agricultural revolution, 1815–1880. Economic

History Review 21(1):62–77Trachtenberg A (1982) The Incorporation of America: Culture and Society in the Gilded

Age. New York: Hill and WangTomich D (1990) Slavery in the Circuit of Sugar: Martinique and the World-economy,

1830–1848. Baltimore: Johns Hopkins University PressVance J E (1988) Transportation and the geographical expression of capitalism. In

E D Genovese and L Hochberg (eds) Geographic Perspectives in History (pp 119–143).Oxford: Basil Blackwell

Walton J (1996) Sociology and Critical Inquiry. New York: WadsworthWallerstein I (1974) The Modern World-System I. New York: Academic PressWallerstein I (1989) The Modern World-System III. San Diego: Academic PressWebb A D (1911) The New Dictionary of Statistics. London: George Routledge and

Sons; New York: E P DuttonWhite R (1991) “It’s Your Misfortune and None of My Own”: A New History of the

American West. Norman: University of Oklahoma PressWilentz S (1984) Against exceptionalism: Class consciousness and the labor

movement, 1790–1920. International Labor and Working Class History 26(Fall):1–24


Wolf E R (1969) Peasant Wars of the Twentieth Century. New York: Harper and RowWolf E R (1982) Europe and the People without History. Berkeley: University of

California PressWoodruff W (1973) The emergence of an international economy. In C Cipolla (ed)

The Fontana Economic History of Europe: The Emergence of Industrial Societies, PartTwo (pp 656–737). New York: Collins/Fontana Books

Zinn H (1980) A People’s History of the United States. New York: Vintage

204 Antipode

Documents

Remaking Work, Remaking Space: Spaces of Production and Accumulation in the Reconstruction of American Capitalism, 1865-1920